>Over the last 14 years of OS development, I've found having a
>debugger in a production kernel to be invaluable (especially when
In my little experience I never debugged the kernel with a debugger. I
never used xkdebugger in the current ikd too even if I maintain it ;).
I think the reason there isn't a debugger in Linus's tree is that passing
the time playing with a debugger may lead to fix the symptom and not the
real bug. I think we should always try to fix bugs by looking C code.
The point is that most of the time we don't need a debugger (as we need in
userspace) to the the backtrace of the fault. And if you get a lockup with
all CPU clied a debugger won't help unless you drive the keyboard irq as a
NMI interrupt. Possible but I don't think you do it now in your patch.
Said that -> but I am quite fashinated by debugger internals and I like to
have it as a _last-resort_ thing in the ikd patch ;).
>While I've no idea what Linus and others reactions will be to including
>kdb in the base kernel release would be, I'll champion it for now, and
>if it is deemed that it doesn't belong there, then I'll be glad to
>integrate it into IKD (It should probably be done for the 2.2 series
Fine. So I really think I'll integrate it into IKD. Really if you want you
can send me a patch against the current ikd I released yesterday against
2.2.8 ;), otherwise I'll do the merge myself from your latest patch
against 2.2.3 sometime soon. Let me know...
I think I won't drop the current debugger of the ikd patch, but maybe I'll
update it soon with the latest version from Dave.
Andrea Arcangeli
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/